Automatic Selection of Basic Level Concepts

Basic Level Concepts (BLC) are those concepts that are frequent and salient; they are neither overly general nor too specific. These BLC are a compromise between two conflicting principles of characterization:

  • to represent as many concepts as possible (abstract concepts)
  • to represent as many distinctive features as possible (concrete concepts).  

We have developed a method for the automatic selection of BLC from WordNet.  We use a very simple method for deriving a small set of appropriate meanings using basic structural properties of WordNet. The program considers:

  • The total number of relations of every synset or just the hyponymy relations.
  • Discard those BLCs that do not represent at least a number of synsets.
  • Optionally, the frequency of the synsets (summing up the frequency of the senses provided by WordNet).

NEW! [2016/09] :

Old version [2009/02] :


This package is distributed under Attribution 3.0 Unported (CC BY 3.0) license. You can find it at


Izquierdo R., Suárez A. and Rigau G. Exploring the Automatic Selection of Basic Level Concepts. Proceedings of the International Conference on Recent Advances on Natural Language Processing (RANLP'07), Borovetz, Bulgaria. September, 2007.

Izquierdo R., Suárez A. and Rigau G. An Empirical Study on Class-based Word Sense Disambiguation. Proceedings of the 12th Conference of the European Chapter of the Association for Computational Linguistics (EACL-09). Athens, Greece. 2009.

Izquierdo R., Suárez A. and Rigau G. Word vs. Class-Based Word Sense Disambiguation.Journal of Artificial Intelligence Research. Volume 54, pages 83-122. 2015.